mlsanityData to 7 Oct 2026

Mirror Descent and the Information Ratio

arXiv 2009.12228 ↗First posted 25 Sep 2020math.OC cs.LG stat.MLcited by 39

Not on the map. Why?

Abstract (arXiv)

We establish a connection between the stability of mirror descent and the information ratio by Russo and Van Roy [2014]. Our analysis shows that mirror descent with suitable loss estimators and exploratory distributions enjoys the same bound on the adversarial regret as the bounds on the Bayesian regret for information-directed sampling. Along the way, we develop the theory for information-directed sampling and provide an efficient algorithm for adversarial bandits for which the regret upper bound matches exactly the best known information-theoretic upper bound.

Key facts. Cited by 39 papers in the map (data through 7 Oct 2026). First posted on arXiv 25 Sep 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 6; 2022: 10; 2023: 10; 2024: 6; 2025: 5; 2026: 2.

39 papers.

Sort:

  1. 2021Feel-Good Thompson Sampling for Contextual Bandits and Reinforcement Learningcited by 65
  2. 2021Reinforcement Learning, Bit by Bitcited by 59
  3. 2022On the Complexity of Adversarial Decision Makingcited by 32
  4. 2022Regret Bounds for Information-Directed Reinforcement Learningcited by 24
  5. 2021Information Directed Sampling for Sparse Linear Banditscited by 20
  6. 2022Deciding What to Model: Value-Equivalent Sampling for Reinforcement Learningcited by 19
  7. 2023Bayesian Design Principles for Frequentist Sequential Learningcited by 17
  8. 2022Contextual Information-Directed Samplingcited by 16
  9. 2021The Value of Information When Deciding What to Learncited by 15
  10. 2022Minimax Regret for Partial Monitoring: Infinite Outcomes and Rustichini's Regretcited by 13
  11. 2023Linear Partial Monitoring for Sequential Decision-Making: Algorithms, Regret Bounds and Applicationscited by 11
  12. 2023An Information-Theoretic Analysis of Nonstationary Bandit Learningcited by 11
27 more
  1. 2024Assouad, Fano, and Le Cam with Interaction: A Unifying Lower Bound Framework and Characterization for Bandit Learnabilitycited by 11
  2. 2023Statistical Complexity and Optimal Algorithms for Non-linear Ridge Banditscited by 7
  3. 2023Improved Bayesian Regret Bounds for Thompson Sampling in Reinforcement Learningcited by 7
  4. 2024Optimistic Information Directed Samplingcited by 7
  5. 2021A Bit Better? Quantifying Information for Bandit Learningcited by 5
  6. 2022Gaussian Imagination in Bandit Learningcited by 5
  7. 2023Incentivizing Exploration with Linear Contexts and Combinatorial Actionscited by 5
  8. 2023Bayesian Reinforcement Learning with Limited Cognitive Loadcited by 4
  9. 2023Synaptic Weight Distributions Depend on the Geometry of Plasticitycited by 4
  10. 2024Understanding Memory-Regret Trade-Off for Streaming Stochastic Multi-Armed Banditscited by 4
  11. 2021Minimax Regret for Bandit Convex Optimisation of Ridge Functionscited by 3
  12. 2022Efficient Algorithms for Learning to Control Bandits with Unobserved Contextscited by 3
  13. 2022A PDE-Based Analysis of the Symmetric Two-Armed Bernoulli Banditcited by 3
  14. 2023Optimal Exploration is no harder than Thompson Samplingcited by 3
  15. 2024Exploration by Optimization with Hybrid Regularizers: Logarithmic Regret with Adversarial Robustness in Partial Monitoringcited by 3
  16. 2024On Bits and Bandits: Quantifying the Regret-Information Trade-offcited by 3
  17. 2022Worst-case Performance of Greedy Policies in Bandits with Imperfect Context Observationscited by 2
  18. 2024Regret Minimization via Saddle Point Optimizationcited by 2
  19. 2022Improved Algorithms for Bandit with Graph Feedback via Regret Decompositioncited by 1
  20. 2025Non-stationary Bandit Convex Optimization: A Comprehensive Studycited by 1
  21. 2025Instance-Dependent Regret Bounds for Nonstochastic Linear Partial Monitoringcited by 1
  22. 2025Sparse Optimistic Information Directed Samplingcited by 1
  23. 2026The Sample Complexity of Multiclass and Sparse Contextual Banditscited by 1
  24. 2023On the Minimax Regret in Online Ranking with Top-k Feedbackcited by 0
  25. 2025On the Problem of Best Arm Retentioncited by 0
  26. 2025Information-directed sampling for bandits: a primercited by 0
  27. 2026Thompson Sampling Is 2-Competitive for Mistakescited by 0